Beyond alignment: Why robotic foundation models need context-aware safety.

Robey, Alexander; Ravichandran, Zachary; Jones, Eliot Krzysztof; Perlo, Jared; Barez, Fazl; Kumar, Vijay; Kolter, J Zico; Hassani, Hamed et al. · Sci Robot · 2026

Where this comes from

Record sourced from PubMed, PMID 42054474.
Also identified by DOI 10.1126/scirobotics.aef2191.
No licence information is recorded for this record.
Because redistribution is not established, this page shows the abstract only. Follow the links below for the full text.

Abstract

Because AI-enabled robots can be tricked into taking unsafe actions, they require layered, context-aware safety guardrails.

Read the full text

View on PubMed · Publisher (DOI)